Transition-regime coverage prediction on METR-LA (test)
84.5Transition-regime CoverageMoirai
Evaluation results
| Method | Links | |
|---|---|---|
| MoiraiCorrection=BMA + ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 84.5 | |
| MoiraiCorrection=BMA, Horizon=h=12, Nominal coverage=90%2026.06 | 84.1 | |
| Chronos-T5Correction=BMA + ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 81.9 | |
| Historical conditional baselineHorizon=h=12, Nominal coverage=90%2026.06 | 81.6 | |
| MoiraiCorrection=R-ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 81.6 | |
| MoiraiCorrection=G-ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 80.9 | |
| Chronos-BoltCorrection=BMA + ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 80.3 | |
| MoiraiCorrection=Native, Horizon=h=12, Nominal coverage=90%2026.06 | 80.3 | |
| Chronos-BoltCorrection=BMA, Horizon=h=12, Nominal coverage=90%2026.06 | 78.7 | |
| Chronos-T5Correction=BMA, Horizon=h=12, Nominal coverage=90%2026.06 | 78.3 | |
| Chronos-T5Correction=G-ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 71.3 | |
| Chronos-T5Correction=R-ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 70.7 | |
| Chronos-BoltCorrection=R-ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 70.2 | |
| Chronos-BoltCorrection=G-ACI, Horizon=h=12, Nominal coverage=90%2026.06 | 69.7 | |
| Chronos-BoltCorrection=Native, Horizon=h=12, Nominal coverage=90%2026.06 | 68.6 | |
| Chronos-T5Correction=Native, Horizon=h=12, Nominal coverage=90%2026.06 | 68.2 |